Vue d'ensemble

  • Missions postés 0

Description de l'entreprise

DeepSeek’s First-generation Reasoning Models

DeepSeek’s first-generation reasoning designs, achieving efficiency comparable to OpenAI-o1 across mathematics, code, and thinking jobs.

Models

DeepSeek-R1

Distilled models

DeepSeek group has actually demonstrated that the thinking patterns of larger models can be into smaller sized models, leading to much better performance compared to the reasoning patterns discovered through RL on small designs.

Below are the designs created via fine-tuning against several thick designs widely used in the research community utilizing thinking information generated by DeepSeek-R1. The evaluation results show that the distilled smaller sized thick models carry out exceptionally well on benchmarks.

DeepSeek-R1-Distill-Qwen-1.5 B

DeepSeek-R1-Distill-Qwen-7B

DeepSeek-R1-Distill-Llama-8B

DeepSeek-R1-Distill-Qwen-14B

DeepSeek-R1-Distill-Qwen-32B

DeepSeek-R1-Distill-Llama-70B

License

The model weights are accredited under the MIT License. DeepSeek-R1 series assistance business use, allow for any adjustments and derivative works, including, but not restricted to, distillation for training other LLMs.